Portfolio Website

Hey there, I am

Akshay

I am a

Data Scientist Machine Learning Engineer Deep Learning Engineer

I turn complex data into actionable insights. With expertise in machine learning and deep learning, I create intelligent models to solve real-world problems.

Featured Projects

Multi-Modal Deepfake Detection

Python, PyTorch, EfficientNet, LSTM, Attention Mechanism, OpenCV

Developed a deep learning model for detecting deepfake videos by integrating visual and audio modalities for enhanced multimedia security.
Leveraged EfficientNet for spatial feature extraction and combined it with LSTM and attention mechanisms to capture spatiotemporal dependencies.
Implemented frame extraction and audio separation pipelines, optimizing the model using PyTorch.
Performed end-to-end preprocessing of video datasets for consistent model input.

LungX: X-Ray Classification

Python, PyTorch, CNN

Built an image classification model using CNN to classify chest X-ray images into normal and abnormal categories.
Implemented advanced data augmentation techniques and fine-tuned hyperparameters to improve model generalization.
Evaluated performance using precision, recall, F1-score, and confusion matrix.
Conducted extensive preprocessing of X-ray datasets, including contrast enhancement and normalization.

Audio Sentiment Analysis

Python, PyTorch, Librosa, CNN, MFCC

Designed a deep learning pipeline using CNN models for classifying sentiments from audio recordings.
Extracted and preprocessed acoustic features using Librosa and implemented a PyTorch-based CNN architecture.
Achieved high sentiment classification accuracy by tuning convolutional layers and dropout rates.
Applied noise filtering and silence removal techniques during preprocessing.